Association of Antibody Responses to the ConservedPlasmodium falciparum Merozoite Surface Protein 5 withProtection against Clinical MalariaRonald Perraut1*, Charlotte Joos1,2, Cheikh Sokhna3, Hannah E. J. Polson2, Jean-Francois Trape3,

Adama Tall4, Laurence Marrama4, Odile Mercereau-Puijalon5, Vincent Richard4, Shirley Longacre2

1 Unite d’Immunologie, Institut Pasteur de Dakar, Dakar, Senegal, 2 Laboratoire de Vaccinologie-Parasitaire, Institut Pasteur, Paris, France, 3 Laboratoire de Paludologie/

Zoologie Medicale, IRD, Dakar, Senegal, 4 Unite d’Epidemiologie, Institut Pasteur de Dakar, Dakar, Senegal, 5 Unite d’Immunologie Moleculaire des Parasites, CNRS URA

2581, Institut Pasteur, Paris, France

Abstract

Background: Plasmodium falciparum merozoite surface protein 5 (PfMSP5) is an attractive blood stage vaccine candidatebecause it is both exposed to the immune system and well conserved. To evaluate its interest, we investigated theassociation of anti-PfMSP5 IgG levels, in the context of responses to two other conserved Ags PfMSP1p19 and R23, withprotection from clinical episodes of malaria in cross-sectional prospective studies in two different transmission settings.

Methods: Ndiop (mesoendemic) and Dielmo (holoendemic) are two Senegalese villages participating in an on-going long-term observational study of natural immunity to malaria. Blood samples were taken before the transmission season (Ndiop)or before peak transmission (Dielmo) and active clinical surveillance was carried out during the ensuing 5.5-month follow-up. IgG responses to recombinant PfMSP5, PfMSP1p19 and R23 were quantified by ELISA in samples from surveys carriedout in Dielmo (186 subjects) and Ndiop (221 subjects) in 2002, and Ndiop in 2000 (204 subjects). In addition, 236 sera fromthe Dielmo and Ndiop-2002 surveys were analyzed for relationships between the magnitude of anti-PfMSP5 response andneutrophil antibody dependent respiratory burst (ADRB) activity.

Results: Anti-PfMSP5 antibodies predominantly IgG1 were detected in 60–74% of villagers, with generally higher levels inolder age groups. PfMSP5 IgG responses were relatively stable for Ndiop subjects sampled both in 2000 and 2002. ADRBactivity correlated with age and anti-PfMSP5 IgG levels. Importantly, PfMSP5 antibody levels were significantly associatedwith reduced incidence of clinical malaria in all three cohorts. Inclusion of IgG to PfMSP1p19 in the poisson regressionmodel did not substantially modify results.

Conclusion: These results indicate that MSP5 is recognized by naturally acquired Ab. The large seroprevalence andassociation with protection against clinical malaria in two settings with differing transmission conditions and stability overtime demonstrated in Ndiop argue for further evaluation of baculovirus PfMSP5 as a vaccine candidate.

Citation: Perraut R, Joos C, Sokhna C, Polson HEJ, Trape J-F, et al. (2014) Association of Antibody Responses to the Conserved Plasmodium falciparum MerozoiteSurface Protein 5 with Protection against Clinical Malaria. PLoS ONE 9(7): e101737. doi:10.1371/journal.pone.0101737

Editor: Kevin K. A. Tetteh, London School of Hygiene and Tropical Medicine, United Kingdom

Received February 13, 2014; Accepted June 11, 2014; Published July 21, 2014

Copyright: � 2014 Perraut et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This work was supported by grants from the Institut Pasteur Foundation, the French Ministry of Research and Technology (Programme VIHPAL/ PAL+),the European Union (EUROMALVAC2 Consortium; project QLRT-2002-01197). HEJ Polson was supported by EUROMALVAC2 funds and Fonds dedies program:fight against Parasitic Diseases (Sanofi Aventis / Ministere de l’Enseignement Superieur et de la Recherche). C Joos was supported by PhD grants from AUF(Agence Universitaire de la Francophonie) and PSF (Pharmacien Sans Frontiere). The funders had no role in study design, data collection and analysis, decision topublish, or preparation of the manuscript.

Competing Interests: The authors confirm that there are no competing interests with this manuscript. Fonds Dedies is a grant from Sanofi/Aventis to InstitutPasteur free for any research topic. Fonds Dedies was a gift to the Institut Pasteur in the context of the French law allowing industry to deduce from their benefitsfunds given for academic research. SANOFI was not aware of the objective of the investigations supported by their grant. The grant from the commercial funderfor Fonds Dedies does not alter the authors’ adherence to PLOS ONE policies on sharing materials.

* Email: perraut@pasteur.sn

Introduction

Plasmodium falciparum malaria is one of the most important

causes of morbidity and mortality worldwide, currently killing over

650,000 people annually, primarily African children under 5 years

old. While scaled up control measures have decreased malaria

morbidity and mortality in many areas of Africa [1], these efforts

are threatened by parasite drug-resistance and anopheles vectors’

insecticide resistance [2,3]. In addition, natural immunity is

waning as a result of reduced exposure to the parasite [4] leaving

endemic populations at increased risk. Development of novel tools

is needed to achieve the objective of control and elimination,

amongst which efficient malaria vaccines.

The protective role of antibodies against blood stage malaria has

been demonstrated using passive immunisation via transfer of

antibodies from hyperimmune African adults to P. falciparumpatients [5,6]. However, it remains unclear which of the many

antibody specificities present in hyperimmune sera are implicated

PLOS ONE | www.plosone.org 1 July 2014 | Volume 9 | Issue 7 | e101737

in protection, information of great relevance for vaccine develop-

ment. One approach to this problem is to investigate relationships

between the antibody response to specific plasmodial antigens and

the immune status of individuals naturally exposed to malaria in

endemic areas.

Clinical symptoms of malaria occur during the blood stage of

the parasitic cycle, during which asexual merozoites invade red

blood cells, multiply intra-cellularly and egress to reinvade new

cells in a cyclical process. Erythrocyte invasion is a rapid, multi-

step process involving a number of merozoite membrane proteins

accessible to immune effectors such as antibodies and complement

[7]. Many merozoite surface proteins (MSPs) are anchored to the

plasma membrane by a C-terminal glyco-lipid moiety (glycosyl-

phosphatidyl-inositol, GPI), often attached to epidermal-growth

factor (EGF)-like domains [7,8].

The first identified and most studied MSP is merozoite surface

protein 1 (MSP1), a 200 kDa protein proteolytically processed to a

conserved C-terminal GPI anchored moiety of around 19 kDa

called MSP1p19 composed of two adjacent EGF-domains [9,10].

Naturally acquired antibodies binding P. falciparum MSP1p19

are major contributors to invasion inhibitory activity present in the

serum of immune adults [11] and are correlated in an age-

independent manner with clinical protection in endemic areas

[12,13,14]. However, MSP1 is only one of several merozoite based

immune targets [15,16] and it is important to identify additional

surface antigens of potential interest for development as vaccine

candidates [17].

One such target of interest is P. falciparum MSP5. The msp5gene codes for a 272-residue protein with a C-terminal EGF-like

domain and a GPI attachment motif [18]. While MSP5 function

in P. falciparum is unknown, it is apparently not critical for

parasite survival since viable knock-out mutants can be isolated

with no apparent growth defect, at least under in vitro culture

conditions [7]. However, MSP5 displays a surprising lack of

population polymorphism in a parasite species renowned for its

immune evasion strategy [8,19,20] and this feature is of particular

interest for a vaccine designed to confer broad cross-strain

protection. Nevertheless, there has been a notable paucity of

epidemiological data monitoring antibody responses to PfMSP5 in

the sera of endemically exposed populations, with only one recent

publication on the subject [21].

The aim of the present study was to evaluate the relationship

between P. falciparum MSP5 antibodies in sera from Senegalese

endemic inhabitants using a baculovirus recombinant PfMSP5,

and the protection status of the same individuals with regards to

clinical malaria episodes and evaluate how this response vis a vis a

response to other low polymorphic Ags already explored in this

setting i.e. MSP1p19 and R23, an infected red blood cell

associated repeat Ag [22]. The study was based on prospective

serological studies carried out in two rural villages with similar

socio-economical levels located 15 km apart in the same district of

southern Senegal. Both villages participate in a longitudinal

follow-up with the same protocol, but differ in transmission

conditions, with holoendemic conditions in Dielmo and mesoen-

demic conditions in Ndiop. We measured the antibody response to

MSP5 in sera collected in Dielmo and Ndiop before the

transmission season, and all participants were monitored for

clinical episodes by active daily case surveillance during the

following 5.5 months [23]. We investigated whether the anti-

MSP5 response was associated with protection against clinical

malaria attacks during that period. Functionality of the antibodies

was assessed using the in vitro assay based on neutrophil antibody-

dependent respiratory bursts induced by opsonized merozoites

(ADRB) [24] was investigated in a subset of samples. In

complement, the stability over time of villagers’ response to

MSP5 and association with protection against malaria episode was

studied through the analysis of 2000 data in Ndiop. The analysis

was done in a multivariate model integrating MSP5 with other

responses against MSP1p19 and R23.

The results provide a compelling case for the interest of

baculovirus PfMSP5 as a blood stage malaria vaccine candidate.

Methods

Study area, study design and proceduresSubjects were recruited in Ndiop and Dielmo, two Senegalese

malaria endemic villages participating in a long-term observational

study on the acquisition and maintenance of natural immunity to

malaria [25,26]. The protocol was approved by the ad hoc Ethics

Committee of the Ministry of Health. Each year, the project was

reviewed by the Conseil de Perfectionnement de l’Institut Pasteur

de Dakar and the assembled village population, and informed

consent was individually renewed by all subjects. Individuals could

withdraw from the study and the follow-up procedure at any time.

The protocol and objectives were carefully explained to the

assembled villagers, and informed consent was obtained from all

participants or their parents or guardians in the presence of an

independent witness and confirmed by signature or by thumbprint

on a voluntary consent form written in both French and Wolof,

the local language.

Villagers were enrolled in the cross-sectional study in July 2002,

before the rainy season ie before peak transmission in Dielmo,

where transmission occurs year round due to a nearby stream, and

before the transmission season in Ndiop (meso-endemic). These

studies involved 186 and 221 healthy villagers, in Dielmo and

Ndiop, respectively, i.e. 67% and 68% of the village population at

that time. In Ndiop, Ab responses were also measured in 204

blood samples collected in July 2000, of which 141 were from

villagers analysed in the 2002 survey. Characteristics of the three

groups are summarized in Table 1, including the proportion of

asymptomatic carriers with microscopically positive peripheral

parasitemia at the time of blood sampling (range 0.5–80

trophozoites per 100 leucocytes). The detection threshold by

experienced microscopists was 2 parasites per microliter i.e. 1

trophozoite per 200 fields of 0,5 ml of thick blood film from the

systematic active follow-up protocol [23,27]. After venous

puncture, plasma and red blood cells were separated by

centrifugation and stored at 220uC.

Active clinical surveillance was carried out during the following

5.5-month transmission period. The protocol included notification

of all febrile episodes and controlled use of anti-malarial drugs by

the medical staff. Each villager was visited daily at home for

clinical surveillance, and blood films were made in case of fever,

and read extemporaneously, as described [13,26,28]. Because of

different endemicity, parasite exposure and immune status, the

definition of a malaria attack differs for Ndiop and Dielmo. In

Ndiop, a clinical episode is defined as an association of symptoms

suggesting malaria; fever (temperature $38uc), history of fever or

diarrhea, with parasitemia .30 trophozoites/100 leukocytes

whatever the age groups. In Dielmo, it is defined as an episode

of fever (temperature .38.5 uC) associated with a parasite density

exceeding an age-dependent pyrogenic threshold determined for

this village. We used the threshold calculated after taking into

account the impact of control measures implemented locally on

malaria epidemiology [4,29,30]. Anti-malarial drugs were admin-

istered by the medical staff following each positive diagnosis of

malaria [4,23,30].

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 2 July 2014 | Volume 9 | Issue 7 | e101737

The cumulative entomological inoculation rate (EIR) was

monitored as described [25]. In Dielmo, malaria transmission

was perennial with a cumulative EIR estimated at 295.5 infective

bites per individual during the clinical follow up period of 2002

(July-Dec). In 2002, the cumulative estimated Ndiop EIR was 17.9

infective bites/individual from the beginning of September to the

end of November, with no transmission in July, August and

December. Two years before, the estimated EIR in Ndiop was

substantially higher i.e. 50.75 infective bites/individual from end

July to October, with no transmission recorded in November and

December.

Antigen and ELISA procedureA synthetic gene coding for PfMSP5 (NF54 allele) was

constructed with a codon usage adapted to baculovirus expression

(manuscript in preparation). The construct encoded a polypeptide

of 247 amino acids from Met-1 to Ile-247 with the C-terminal 25

amino acids mediating GPI modification deleted to allow protein

secretion and replaced with a hexa-histidine tag to facilitate

purification. Serine codons at positions S-80, S-99 and S-124 were

replaced by alanine codons to prevent N-glycosylation. Recom-

binant baculovirus was produced using the pVL1393 transfer

vector as previously described [31]. Recombinant PfMSP5 was

secreted following infection of Trichoplusia ni insect cells (High

Five, Invitrogen) with recombinant baculovirus, and purified by

metalloaffinity chromatography using Talon, as previously de-

scribed [32]. The baculovirus expression system has been shown to

ensure optimal reproduction of conformational epitopes including

EGF domains [10]. Recombinant PfMSP5 was diluted in PBS and

used to coat Immulon-4 plates at a concentration of 0.5 mg mL21.

IgG responses were quantified by ELISA in duplicate plasma

samples diluted 1:200 as previously described [13,14], IgG

subclasses were determined using human sub-class specific mouse

mAbs and peroxidase-labeled goat anti-mouse IgG (1:2000) from

Sigma Chemicals (St Louis, Mo) after calibration for optimal

concentrations [33,34,35]. A pool of 25 sera from clinically

immune adults living in Dielmo and a pool of European and/or

African non-immune sera were included as positive and negative

controls, respectively, in each assay as standards for plate

comparability. Results were expressed as OD ratio = OD sample

/ OD naive serum pool [13,35]. Sera showing an OD ratio .2

corresponding to the signal of naive controls + 2 SD were

considered sero-positive for prevalence calculations.

The Ags and ELISA test for IgG responses to PfMSP1p19 and

to R23 was carried out as already described [13,14,22,36,37].

Chemiluminescence monitoring of ADRB by neutrophilsin the presence of merozoites

The ADRB assay has been detailed elsewhere [24]. Briefly,

polymorphonuclear cells (PMNs) were harvested from pooled

blood samples of 6–7 healthy donors by layering onto Ficoll-

Hystopaque (density 1.077, Sigma) and collected after centrifuga-

tion 30 min at 400xg at the Ficoll-Hystopaque-RBC interface.

After differential lysis of residual RBC using NH4Cl, 0.8 g.L21

buffer, PMN were washed twice with Hank’s balanced salt solution

(HBSS), enumerated using trypan blue, and resuspended in PBS at

1–5.107 cells mL21.

Chemiluminescence was measured using opaque 96-well plates

(Berthold), and a MicroLumat Plus 96 luminometer (Berthold).

Merozoite pellets (40 mL) were incubated with 10 mL of test or

control sera for at least 30 min at 37uC. Immediately following

addition of PMN (100 mL) and isoluminol (100 mL of 1:100

dilution in PBS of 4 mg.mL21 stock in DMSO) plate reading

started and continued for 1 h. To facilitate rapid handling, only

40–50 wells per plate were used, with the hyperimmune serum

(HIS) internal control systematically deposited in the first and last

wells.

Data are presented as a standardized activity index of ADRB

calculated as:

ADRB index~

rlu maximum sample=rlu maximum HISð Þx1000

where rlu maximum HIS was the average of readout from the first

and last wells on each plate. Only experiments in which the rlu

maximum HIS was $100 (i.e. $6x background) were included in

the analyses. An additional internal control with the same

individual positive serum was included in each run.

Statistical analysisAntibody levels and growth and/or invasion inhibition in

different groups were compared using the Wilcoxon signed rank

test and the Spearman rank correlation test for non-normally

distributed paired data. P values ,0.05 were considered signifi-

cant. A Poisson regression model was used to analyze the

relationship between antibody response(s) and incidence of malaria

Table 1. Characteristics of populations analyzed in the study.

Dielmo Year 2002 Ndiop Year 2002 Ndiop Year 2000

Number of villagers 186 221 204

sex ratio F/M 99/87 119/102 97/107

Mean age 28.6 24.2 24

Median age [range] 22.8 [3.4–80.5] 18.2 [3.4–76.9] 18.2 [3.6–75]

No Indiv. Hb AA/AS/AC 160/14/4 186/30/5 164/36/4

% individuals parasitemic* 45.9% 9.6% 18.1%

cumulative EIR** 295.5 17.9 50.7

No Indiv. included for clin. attacks 163 203 191

Overall No of clinical attacks 51 199 275

*Individuals with positive blood smear on the day of blood withdrawal.**Cumulative Entomological Inoculation Rate measured during the 5 months follow-up.doi:10.1371/journal.pone.0101737.t001

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 3 July 2014 | Volume 9 | Issue 7 | e101737

attacks during the 5.5-month follow-up period. For the analysis, a

P. falciparum malaria episode was defined as the presence of fever

or symptoms suggesting malaria associated with .30 P.falciparum trophozoites/100 leukocytes, as determined by the

same experienced microscopist. Malaria attacks were considered

independent if separated by .15 days. For each villager, the

follow-up time was calculated as the number of days actually spent

in the village during the 5.5-month study and those absent more

than 30 days during this period were excluded from the malaria

incidence analysis. For individuals who received anti-malarial

treatment following a malaria attack, a period of 15 days after the

diagnosed attack was excluded from the days at risk calculation.

The numbers of individuals and clinical episodes included in the

calculations for each cohort are noted in Table 1. Age stratification

was based on the age distribution of the parasitologic and clinical

data available for Ndiop and Dielmo during the previous

approximately 10 years of longitudinal follow up of both

populations ie ,15;15–30; .30 and ,7;7–15; .15 years old

age groups for Ndiop and Dielmo, respectively [14,29,30,38],

during which endemicity did not markedly change in each village

[23,28]. First-level interactions between variables were checked

and included in the model when significant. The antibody level

stratification was determined using Aikake’s information criterion.

Statistical analyses were performed with Egret 3.01 (Cytel), R and

Statview 5.0 (SAS Institute) softwares.

Results

Seroprevalence and antibody levels to PfMSP5,PfMSP1p19 and R23

Table 2 shows that IgG reacting with baculovirus PfMSP5 was

detected in 74% and 65% of villagers in Dielmo and Ndiop,

respectively. Prevalence in Ndiop increased from 58% in the

,15 y age-group to 70% in the .15 y group without reaching

statistical significance (P = 0.058). In Dielmo, prevalence was high

even in the youngest age group (,7 y). There was no significant

difference in overall levels of anti-PfMSP5 IgG response (mean

OD/median and OD ratios) between the two villages

The age distribution of clinical accesses and anti-PfMSP5 IgG

responses in the two 2002 follow up studies is shown in Figure 1 (A

and C). Inserts show the age-stratified clinical malaria incidence

(panels a and c) and the anti-PfMSP5 IgG responses using the

same clinical age-stratification (panels b and d). Significant

differences were observed between the 7–14 y and .15 y groups

in Dielmo, as well as between the ,15 and $15–29 y groups in

Ndiop. In Ndiop, there was a general trend towards higher anti-

PfMSP5 responses in older age groups (P,0.05). In Dielmo, the

higher responses observed in the ,7 y group compared to the 7–

14 y old children is possibly relate to substantial and recent

exposure of individuals (134 positive smears [100% of individuals]

and 47 clinical episodes [87% of individuals] in the 3 months

before July), although this may reflect also a small sample size

effect of the youngest age group (n = 19). A significant relationship

between age and IgG response to PfMSP5 (P,0.01) was observed

in Ndiop only, albeit with a low correlation coefficient (0.2–0.25).

Asymptomatic parasitemia on the day of sampling was

associated with higher anti-PfMSP5 IgG only for the individuals

from Ndiop (P,0.001). No correlations between antibody

responses and gender or hemoglobin phenotype (P.0.05) were

observed in either village.

In Table 3 are shown Levels and prevalence of IgG responses

against PfMSP1p19 in the three settings, showing similar high

levels of responses (mean ODratio above 7.9) and high prevalence

in all age groups. There was a significant correlation between age

of individuals and IgG to PfMSP1p19 in the 3 cohorts (P,1023,

correlation coefficient around 0.4). IgG response to R23, was

substantially lower in the 2 villages in 2002 (mean OD/ODratio of

0.22/2.3 and 0.21/2.7) with a prevalence of 24% and 30% in

Dielmo and Ndiop, respectively. Ab responses to R23 correlated

with age in Ndiop only (P,1023, correlation coefficient = 0.27).

Comparison of anti-PfMSP5 antibodies in NdiopIn Ndiop, it was possible to retrospectively evaluate the IgG

responses to PfMSP5 of 204 villagers in a previous sampling

conducted in July 2000, including 141 villagers who participated

in the 2002 study. It was thus possible to compare the response in

this sub-group of 141 villagers participating in both the 2000 and

2002 studies. Both seroprevalence (60% vs 71%) and antibody

levels (ODratios = 4.1 vs 4.4) showed very similar results in the

2000 and 2002 surveys with no significant difference by the

Wilcoxon rank test.

Of the 141 individuals evaluated both years, 31 (22%) having a

large age-distribution (5.2 to 50.3 years old in 2000) were negative

in both surveys, whereas 74 others (52%) scored positive at both

time points, showing comparable elevated levels of Ab with mean

ODratios of 5.9 and 6.5 in 2000 and 2002, respectively In Figure 2

illustrating the degree of individual variation, panel A shows Ab

responses for the 105 villagers with unchanged profiles in 2000

and 2002, and panel B shows responses for the 36 villagers scoring

differently in the two studies, with 26 becoming positive between

2000 and 2002 while 10 were negative in 2002 after being positive

in 2000.

Table 2. Prevalence and levels of antibody responses to PfMSP5 in the three cohorts.

Ab responses against PfMSP5

Dielmo Year 2002 Ndiop Year 2002 Ndiop Year 2000

Mean OD/median-[range] 0.41/0.32[0.01–1.7] 0.27/0.16 [0.01–1.46] 0.38/0.21 [0.01–2.63]

Mean ODratio/median-[range] 4.3/3.3 [1–17.1] 4.3/2.7 [1–22.8] 3.9/2.4 [1–23.2]

Prevalence of responders* 74% 65% 60%

Prevalence in younger individuals** 79% (n = 19) 58% (n = 85) 49% (n = 80)

Prevalence in older individuals 74% (n = 167) 70% (n = 136) 70% (n = 124)

*Individuals with positive IgG response = ODratio.2 ie . OD of naive control + 2SD.**Younger individuals : in Dielmo ie ,7 years old; in Ndiop ie ,15 years old.doi:10.1371/journal.pone.0101737.t002

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 4 July 2014 | Volume 9 | Issue 7 | e101737

Isotype distribution of anti-PfMSP5 IgGThe isotype of IgG antibodies to PfMSP5 was investigated using

a set of 30 positive sera with a similar age-distribution as the

cohorts (mean age 24.5 y, median = 18.6 y, range 7.9–72.6 y).

Anti-PfMSP5 antibodies were predominantly IgG1 with only low

levels of other subclasses (Figure 3). IgG1 levels were unrelated to

age, gender, hemoglobin type or microscopically positive, asymp-

tomatic parasitemia at the time of blood sampling (P.0.05).

Association of ADRB activity with age and anti-PfMSP5antibodies

The antibody-dependent respiratory burst (ADRB) assay

measures the capacity of antibodies to opsonize merozoites and

Figure 1. Age-related distribution of clinical episodes and IgG responses to PfMSP5. Distribution of clinical episodes and ELISA OD ratiosreflecting IgG responses to PfMSP5 are plotted according to the age of individuals in Dielmo (A and B) and in Ndiop (C and D) in 2002. In captions areshown age-group distribution of number of clinical episodes in Dielmo (a) and Ndiop (c) and mean ODratio of anti-MSP5 IgG responses in Dielmo (b)and Ndiop (d). Groups were based on distribution of parasitologic and clinical data defined for both populations ie the ,7, 7–14 and .15 y age-groups in Dielmo, and for the ,15, 15–29 and .30 y age-groups in Ndiop. Brackets and asterisk indicate significant age-related differences (P,0.05)between levels of Ab responses.doi:10.1371/journal.pone.0101737.g001

Table 3. Prevalence and levels of antibody responses to PfMSP1p19 in the three cohorts.

Ab responses against PfMSP1p19

Dielmo Year 2002 Ndiop Year 2002 Ndiop Year 2000

Mean OD/median-[range] 0.79/0.55[0.01–1.7] 0.86/0.80 [0–1.96] 0.94/0.79 [0.0–2.69]

Mean ODratio/median-[range] 7.9/5.3 [1–16.9] 6.0/5.5 [1–18.0] 7.9/6.7 [1–22.7]

Prevalence of responders* 73% 74% 78%

Prevalence in younger individuals** 68% (n = 19) 60% (n = 85) 65% (n = 80)

Prevalence in older inviduals 74% (n = 167) 83% (n = 136) 87% (n = 124)

*Individuals with positive IgG response = ODratio.2 ie . OD of naive control + 2SD.**Younger individuals : in Dielmo ie ,7 years old; in Ndiop ie ,15 years old.doi:10.1371/journal.pone.0101737.t003

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 5 July 2014 | Volume 9 | Issue 7 | e101737

induce a neutrophil respiratory burst, which is quantified by

chemiluminescence. Because of its dependence on relatively intact

merozoites [24], the assay is primarily a measure of functional

activity of antibodies reacting with merozoite surface antigens. It

was thus of interest to investigate the relationship between anti-

PfMSP5 antibody levels and ADRB activity. Samples from cross

sectional studies were randomly selected but matched with the

age-group distribution of individuals from the 2002 studies.

Altogether, 236 sera were tested for ADRB activity, including

120 from Dielmo (mean age 25.2; 3.4–80.5 y) and 116 from Ndiop

(mean age 22.9; 3.9–76.9 y). In addition, 93 sera from the Ndiop

2000 sampling (mean age 22.9; 3.9–76.9 y) were assayed for

ABRB activity. ADRB readout was expressed as a standardized

value relative to a control pool of hyperimmune sera (ADRB

index) [24]. Geometric mean values for the ADRB index were 255

(range 40–958), 201 (range 51–1721) and 504 (range 82–1809)

respectively for the three set of samples from Dielmo-2002, Ndiop-

2002 and Ndiop-2000, respectively.

In the three surveys, age and ADRB responses were significantly

correlated: P,0.001 rho = 0.53; 0.41 and 0.44 for Dielmo-2002,

Ndiop-2002 and Ndiop-2000, respectively, and ADRB indexes

were significantly lower in the youngest age group (P,0.05).

ADRB indexes and PfMSP5 antibody responses were positively

correlated (P,0.001 rho = 0.33, 0.40 and 0.29 for Dielmo-2002,

Ndiop-2002 and Ndiop-2000, respectively). When anti-PfMSP5

IgG OD ratios were stratified into negative (ODratio ,2), median

(ODratio 2–4) and high values (ODratio.4), they were positively

associated with significantly different ADRB indexes in all cases

(P = 0.01), suggesting a potential contribution of anti-PfMSP5

antibodies to this functional parameter (Figure 4A).

Association of PfMSP5 antibodies with reduced incidenceof clinical episodes

ADRB indexes were related to protection against clinical

malaria during the 5.5 month follow-up. When stratifying into

three clinical group, (0 vs 1–2 vs .2 cumulative clinical episodes),

there was an inverse relationship with ADRB indexes (Figure 4B).

ADRB indexes decreased as the number of clinical episodes

increased in Dielmo-2002 and Ndiop-2002 (P,0.01) as well as in

Ndiop-2000 (data not shown). This association was independent of

age by the chi square test of independence (P,0.01).

The relationship between the anti-PfMSP5 response and

incidence of clinical malaria during the 5.5-month follow-up

period was analyzed using an age-adjusted Poisson regression

Figure 2. Longitudinal comparison of anti-PfMSP5 antibodies in Ndiop. Anti-PfMSP5 IgG responses for the 141 Ndiop villagers participatingto two cross-sectional studies are shown. Ab responses were measured in both 2000 (light gray) and 2002 (dark grey). A box-whisker plot showslevels of Ab responses of individuals with the same profile ie positive or negative responses for both years (A). In part B are plotted Ab responses of36 sera from individuals with discordant anti-PfMSP5 response in the two periods. Brackets indicate significant different levels between the twosubgroups of villagers present in the two studies. The individuals with discordant responses from both surveys showed significantly lower Ab levelsthan those showing concordant responses (P,0.001).doi:10.1371/journal.pone.0101737.g002

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 6 July 2014 | Volume 9 | Issue 7 | e101737

model for each survey. Ab responses to R23 did not correlate with

protection and were not included in the final multivariate model.

In Dielmo-2002, clinical episodes were significantly associated

with age (7–14 vs $15 y, RR1 = 8.43, P,0.001; 2–6 vs $15 y,

RR2 = 87.54, P,0.001), but not hemoglobin phenotype or

positive parasitemia at sampling. Dichotomization of the anti-

PfMSP5 response based on negative vs positive sera was the best

model for this cohort, showing a significant association with

protection in the age-adjusted analysis (RR = 0.53 [CI

95% = 0.30–0.95]; P = 0.03).

In an age-adjusted model including both anti-PfMSP1 response

(RR = 0.98, 95%CI [0.88-1.08]) and anti-PfMSP5 response

(RR = 0.42, 95%CI [0.23–0.77]), only the anti-PfMSP5 responses

were correlated with clinical malaria (P,0.01).

For the Ndiop-2002 cohort, variables correlated with clinical

malaria were: (i) age (15–29 vs $30 y, RR1 = 2.75, P = 0.004; 2–

14 vs $30 y, RR2 = 13.10; P,0.001); (ii) hemoglobin phenotype

(AA vs AS, RR = 0.44; P = 0.004); (iii) positive parasitemia at

sampling (RR = 1.56, P = 0.012) and (iv) the continuous anti-

PfMSP5 response (OD ratios) in an age-adjusted model

(RR = 0.95 per unit lost; P = 0.002). In an age-adjusted model

including both anti-PfMSP1 response (RR = 0.96, 95%CI [0.93–

0.99]) and anti-PfMSP5 response (RR = 0.97, 95%CI [0.93–

1.01]), only the anti-PfMSP1 response was correlated with clinical

malaria (P-value = 0.04). Analysis of the Ndiop-2000 cohort

showed similar results in age-adjusted model not including anti-

PfMSP1 response. However, according with the age-adjusted

model including both anti-PfMSP1 (RR = 0.97, 95%CI [0.95–

0.99]) response and anti-PfMSP5 response (RR = 0.91, 95%CI

[0.87–0.96]), showed that both anti-PfMSP1 and anti-PfMSP5

responses were correlated with clinical malaria (P-value = 0.02 and

P,0.001, respectively).

Figure 5 shows the temporal incidence of number of malaria

attacks over the 5.5-month follow-up period. Very consistent

results have been obtained in the two settings and the two study

periods.

Indeed, in both villages and age groups, a high OD ratio was

associated with fewer clinical episodes through the 5.5 month

follow-up period. These results are particularly marked in the age

groups and study periods with higher malaria incidence, i.e. in the

youngest most susceptible age group of both villages and in Ndiop

2000 (year-2) where transmission level was higher than in 2002.

The only exception is observed in the oldest age group in Ndiop

2000 where the very low incidence can explain the difference with

the other results (only 2 clinical malaria episodes per 1000 person-

days for the $30 y age group).

Discussion

The present study aimed to investigate the value of PfMSP5 as a

blood stage malaria vaccine candidate by analyzing possible

associations between naturally acquired antibodies to this antigen

in endemic populations and susceptibility to clinical episodes of P.falciparum malaria. PfMSP5 has been relatively neglected in the

literature compared to other P. falciparum MSP vaccine

candidates such as PfMSP1, PfMSP2 and MSP3 or AMA-1 all

of which are polymorphic and moreover induce allele-specific

responses [39,40,41]. PfMSP5 appears to be relatively unique

among blood stage surface antigens in exhibiting virtually no

polymorphism. One hypothesis accounting for this remarkable

conservation would be that PfMSP5 remains somehow undetected

and as such unselected by naturally acquired antibodies. Clearly

with 60–74% responders in this study, PfMSP5 does not escape

the attention of the immune system. Prevalence was high and

antibody levels were substantial. This indicates that other factors –

possibly strong constraints on function- must account for the

strong conservation of PfMSP5.

Two features of the approach used here were particularly

important for maximizing the in vivo relevance of the results

obtained. First, the recombinant PfMSP5 antigen used in the study

was produced in the baculovirus expression system to optimize its

resemblance to the native homologue, especially regarding

conformational epitopes in the EGF domain and elsewhere in

the molecule. Previous work with recombinant MSP1p19 has

shown that this system appears to be particularly competent in

achieving this objective [9,10]. Nevertheless to prevent undesirable

N-glycosylation the serine codons at three potential N-glycosyla-

tion sites were changed to alanine. A previous study on PfMSP5

antibody recognition used an E. coli recombinant antigen [21],

which may not optimally reproduce some native conformational

epitopes. This could account for some differences in results from

the two studies relating to prevalence (39%–50% vs 60–74% here)

and IgG isotype (IgG1 and/or IgG3 vs exclusively IgG1 here).

Secondly, the study was carried out in two endemic Senegalese

villages with well-defined seasonal (Ndiop) and year-round

(Dielmo) transmission, which have participated in a long-term

cross-sectional prospective study with intensive follow-up for over

15 years. The study thus could rely on access to antisera from

three different large cohorts of endemic donors and the protocol

involved active case detection over a long 5.5-month follow-up

with daily medical visits to each participant and the use of efficient

diagnostic criteria to distinguish fevers due to malaria from other

potential causes. These various factors provided the statistical

power underlying many of the findings presented here.

Figure 3. IgG isotype of anti-PfMSP5 responses. A set of 30positive sera from all age groups were analyzed for anti-PfMSP5 IgGsub-classes as described in Methods. OD ratio results are plotted as barcharts + SD.doi:10.1371/journal.pone.0101737.g003

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 7 July 2014 | Volume 9 | Issue 7 | e101737

IgG responses to PfMSP5 generally increased with age,

although the youngest Dielmo age group had substantially high

magnitudes, possibly due to the low number of group members

and/or transient elevated responses by some susceptible individ-

uals having experienced recent infection at this holoendemic site

with frequent infective bites (a mean 295.5 infective bites for

Dielmo-2002 compared to 17.9 and 50.75 for Ndiop in 2002 and

2000, respectively). The general trend associating older individuals

with augmented antibody responses is a frequent observation with

many plasmodial antigens in endemic areas, which likely reflects

the probability of increased cumulative parasite exposure with age.

It has been observed in this study in a very similar manner for IgG

responses to PfMSP1p19 in all cohorts, and to a lesser extent for

R23 in Ndiop only.

Anti-PfMSP5 responses were surprisingly stable in prevalence

and magnitude over two years for the group of 141 individuals

who were members of both the 2000 and 2002 Ndiop studies, the

main difference being the 26% who became sero-positive with low

OD levels during this interval. Since sampling occurred prior to

the transmission season before any boosting by exposure to

parasites, it is likely that the observed anti-PfMSP5 antibody levels

persist for at least 6–7 months between transmission seasons, and

may not change appreciably in the overall population from one

season to the next. Thus, anti-PfMSP5 antibodies do not appear to

be short-lived as suggested elsewhere [21], although the latter

observation may pertain particularly to transient increases due to

active infection.

When a cohort of 30 positive anti-PfMSP5 responder sera was

analyzed for IgG isotype levels, IgG1 was by far the predominant

species with only small amounts of IgG2, IgG3 or IgG4. The

previous study measuring anti-PfMSP5 IgG isotype responses

seemed to implicate also IgG3 [21], but it is difficult to compare

results from the two studies because of differences in recombinant

antigen used to measure responses, cohorts, and readout

(prevalence versus levels). The IgG1 and IgG3 ‘‘cytophilic’’

isotypes couple adaptive and innate immune functions via

Fcgamma receptors (FcgR) expressed on cellular effectors. In

particular, merozoites opsonized with specific IgG1 and/or IgG3

are targets for antibody-dependent phagocytosis by neutrophils,

which are highly professional phagocytes [42,43]. IgG1 and/or

IgG3 opsonized merozoites are also able to elicit respiratory bursts

by neutrophils, an activity that has been correlated with protection

from clinical episodes of P. falciparum malaria [24]. Indeed, we

have shown here a significant positive association between

stratified anti-PfMSP5 antibody levels and antibody-dependent

respiratory burst (ADRB) activity by neutrophils in vitro,

suggesting that PfMSP5 antibodies might be a component of this

mechanism. In this study, a similar relationship between ADRB

and IgG to baculovirus derived PfMSPp19 Ag was evidenced. We

observed in 2002 a significant relationship of ADRB with clinical

protection in an age-adjusted manner in the two villages [24], and

approximately 30% of ADRB activity has been found related to

PfMSPp19 Ag after selective depletion of immune sera (unpub-

lished data). Such a functional in vitro potential capacity of anti

Figure 4. Relationship between ADRB functional antibody responses and anti-PfMSP5 IgG responses or malaria episodes. ADRBindexes were determined for 236 sera (120 for Dielmo-2002, 116 for Ndiop-2002) as described in Methods and compared (A) between individualsstratified by anti-PfMSP5 responses and (B) between individuals stratified by clinical outcome. In A, anti-PfMSP5 responses expressed as OD ratios ,2for negative (light grey bars), 2–4 for moderate (dark grey bars) and .4 for strong (black bars). In B, mean ADRB values are plotted by stratifiedclinical outcome: no (light grey), one or two (dark grey) and more than 2 clinical malaria episodes (black) during the 5.5 month follow up. Bracket andasterisks indicate significant differences (P,0.05).doi:10.1371/journal.pone.0101737.g004

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 8 July 2014 | Volume 9 | Issue 7 | e101737

Figure 5. Association between anti-PfMSP5 IgG responses and cumulative incidence of clinical malaria episodes in different agegroups. The cumulative monthly incidence of clinical attacks across the 5.5-month follow up was determined for individuals with dichotomizednegative i.e. OD ratio #2 (filled symbols, solid lines) vs positive responders, i.e. OD ratio .2 (open symbols, dashed lines) against anti-PfMSP5 IgG foreach cohort. Young individuals (circles) vs older (triangles) are shown for Dielmo-2002 (A: ,7yo vs $7yo; EIR = 295.5), for Ndiop-2002 (B: ,15yo vs$15yo; EIR = 17.9) and for Ndiop-2000 (C; EIR = 50.7).doi:10.1371/journal.pone.0101737.g005

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 9 July 2014 | Volume 9 | Issue 7 | e101737

PfMSP5 Ab could parallel /complement PfMSP1p19 IgG and

warrants to be further explored.

In two different transmission settings, anti-PfMSP5 IgG was

correlated with protection from clinical episodes of malaria. This

was particularly apparent in the youngest, most vulnerable age

groups of each cohort experiencing the overwhelming majority of

episodes. However, it was also evident in the intermediate 15–29 y

group in the Ndiop-2000 study (data not shown), with a high

incidence of clinical cases probably because transmission was

particularly efficient that year. The older groups as expected

generally had too few cases to permit reliable comparisons.

This is the first epidemiological study in which the PfMSP5

antibody response has been investigated in depth in two settings

studied with the same active case detection system using a

relatively long-term intensive follow-up protocol. The results

provide significant support for the interest of PfMSP5 to be

included in a multitarget P. falciparum vaccine candidate, one of

very few with so little polymorphism, potentially offering the

prospect of sustained protection against this accomplished

antigenic chameleon.

Acknowledgments

We are very grateful for technical assistance from B. Diouf with ELISAs

and parasite culture and from S. Corre for ADRB assays. We thank Pr A.

Dieye for constant support of this work. We are particularly indebted to the

villagers from Ndiop and Dielmo who participated in this study.

Author Contributions

Conceived and designed the experiments: RP SL LM OMP VR.

Performed the experiments: RP CJ HEJP. Analyzed the data: RP LM

VR AT CS JFT. Contributed reagents/materials/analysis tools: RP SL

LM VR AT CS JFT OMP. Wrote the paper: RP SL LM OMP VR.

References

1. WHO (2012) WHO World Malaria Report 2012. Available: http://

wwwwhoint/malaria/publications/world_malaria_report_2012/en/.

2. Dondorp AM, Nosten F, Yi P, Das D, Phyo AP, et al. (2009) Artemisinin

resistance in Plasmodium falciparum malaria. N Engl J Med 361: 455–467.

3. Noranate N, Durand R, Tall A, Marrama L, Spiegel A, et al. (2007) Rapid

dissemination of Plasmodium falciparum drug resistance despite strictly

controlled antimalarial use. PLoS One 2: e139.

4. Trape JF, Tall A, Diagne N, Ndiath O, Ly AB, et al. (2011) Malaria morbidity

and pyrethroid resistance after the introduction of insecticide-treated bednets

and artemisinin-based combination therapies: a longitudinal study. Lancet Infect

Dis.

5. Bouharoun-Tayoun H, Attanath P, Sabchareon A, Chongsuphajaisiddhi T,

Druilhe P (1990) Antibodies that protect humans against Plasmodium

falciparum blood stages do not on their own inhibit parasite growth and

invasion in vitro, but act in cooperation with monocytes. J Exp Med 172: 1633–

1641.

6. Cohen S, McGregor IA, Carrington S (1961) Gamma-globulin and acquired

immunity to human malaria. Nature 192: 733–737.

7. Cowman AF, Crabb BS (2006) Invasion of red blood cells by malaria parasites.

Cell 124: 755–766.

8. Riley EM, Stewart VA (2013) Immune mechanisms in malaria: new insights in

vaccine development. Nat Med 19: 168–178.

9. Chitarra V, Holm I, Bentley GA, Petres S, Longacre S (1999) The crystal

structure of C-terminal merozoite surface protein 1 at 1.8 A resolution, a highly

protective malaria vaccine candidate. Mol Cell 3: 457–464.

10. Pizarro JC, Chitarra V, Verger D, Holm I, Petres S, et al. (2003) Crystal

structure of a Fab complex formed with PfMSP1-19, the C-terminal fragment of

merozoite surface protein 1 from Plasmodium falciparum: a malaria vaccine

candidate. J Mol Biol 328: 1091–1103.

11. O’Donnell RA, de Koning-Ward TF, Burt RA, Bockarie M, Reeder JC, et al.

(2001) Antibodies against merozoite surface protein (MSP)-1(19) are a major

component of the invasion-inhibitory response in individuals immune to

malaria. J Exp Med 193: 1403–1412.

12. Dodoo D, Aikins A, Kusi KA, Lamptey H, Remarque E, et al. (2008) Cohort

study of the association of antibody levels to AMA1, MSP119, MSP3 and

GLURP with protection from clinical malaria in Ghanaian children. Malar J 7:

142.

13. Perraut R, Marrama L, Diouf B, Fontenille D, Tall A, et al. (2003) Distinct

surrogate markers for protection against Plasmodium falciparum infection and

clinical malaria identified in a Senegalese community after radical drug cure.

J Infect Dis 188: 1940–1950.

14. Perraut R, Marrama L, Diouf B, Sokhna C, Tall A, et al. (2005) Antibodies to

the conserved C-terminal domain of the Plasmodium falciparum merozoite

surface protein 1 and to the merozoite extract and their relationship with in vitro

inhibitory antibodies and protection against clinical malaria in a Senegalese

village. J Infect Dis 191: 264–271.

15. McCarthy JS, Marjason J, Elliott S, Fahey P, Bang G, et al. (2011) A phase 1

trial of MSP2-C1, a blood-stage malaria vaccine containing 2 isoforms of MSP2

formulated with Montanide(R) ISA 720. PLoS One 6: e24413.

16. Roussilhon C, Oeuvray C, Muller-Graf C, Tall A, Rogier C, et al. (2007) Long-

term clinical protection from falciparum malaria is strongly associated with IgG3

antibodies to merozoite surface protein 3. PLoS Med 4: e320.

17. Richards JS, Arumugam TU, Reiling L, Healer J, Hodder AN, et al. (2013)

Identification and prioritization of merozoite antigens as targets of protective

human immunity to Plasmodium falciparum malaria for vaccine and biomarker

development. J Immunol 191: 795–809.

18. Marshall VM, Tieqiao W, Coppel RL (1998) Close linkage of three merozoite

surface protein genes on chromosome 2 of Plasmodium falciparum. Mol

Biochem Parasitol 94: 13–25.

19. Polson HE, Conway DJ, Fandeur T, Mercereau-Puijalon O, Longacre S (2005)

Gene polymorphism of Plasmodium falciparum merozoite surface proteins 4 and

5. Mol Biochem Parasitol 142: 110–115.

20. Wu T, Black CG, Wang L, Hibbs AR, Coppel RL (1999) Lack of sequence

diversity in the gene encoding merozoite surface protein 5 of Plasmodiumfalciparum. Mol Biochem Parasitol 103: 243–250.

21. Woodberry T, Minigo G, Piera KA, Hanley JC, de Silva HD, et al. (2008)

Antibodies to Plasmodium falciparum and Plasmodium vivax merozoite surface

protein 5 in Indonesia: species-specific and cross-reactive responses. J Infect Dis

198: 134–142.

22. Bonnefoy S, Guillotte M, Langsley G, Mercereau-Puijalon O (1992) Plasmodiumfalciparum: characterization of gene R45 encoding a trophozoite antigen

containing a central block of six amino acid repeats. Exp Parasitol 74: 441–451.

23. Trape JF, Rogier C, Konate L, Diagne N, Bouganali H, et al. (1994) The

Dielmo project: a longitudinal study of natural malaria infection and the

mechanisms of protective immunity in a community living in a holoendemic

area of Senegal. Am J Trop Med Hyg 51: 123–137.

24. Joos C, Marrama L, Polson HE, Corre S, Diatta AM, et al. (2010) Clinical

protection from falciparum malaria correlates with neutrophil respiratory bursts

induced by merozoites opsonized with human serum antibodies. PLoS One 5:

e9871.

25. Fontenille D, Lochouarn L, Diatta M, Sokhna C, Dia I, et al. (1997) Four years’

entomological study of the transmission of seasonal malaria in Senegal and the

bionomics of Anopheles gambiae and A. arabiensis. Trans R Soc Trop Med Hyg

91: 647–652.

26. Trape JF, Rogier C (1996) Combating malaria morbidity and mortality by

reducing transmission. Parasitol Today 12: 236–240.

27. Trape JF (1985) Rapid evaluation of malaria parasite density and standardiza-

tion of thick smear examination for epidemiological investigations. Trans R Soc

Trop Med Hyg 79: 181–184.

28. Trape JF, Tall A, Sokhna C, Ly AB, Diagne N, et al. (2014) The rise and fall of

malaria in a west African rural community, Dielmo, Senegal, from 1990 to 2012:

a 22 year longitudinal study. Lancet Infect Dis 14: 476–488.

29. Rogier C, Commenges D, Trape JF (1996) Evidence for an age-dependent

pyrogenic threshold of Plasmodium falciparum parasitemia in highly endemic

populations. Am J Trop Med Hyg 54: 613–619.

30. Roucher C, Rogier C, Dieye-Ba F, Sokhna C, Tall A, et al. (2012) Changing

malaria epidemiology and diagnostic criteria for Plasmodium falciparum clinical

malaria. PLoS One 7: e46188.

31. Holm I, Nato F, Mendis KN, Longacre S (1997) Characterization of C-terminal

merozoite surface protein-1 baculovirus recombinant proteins from Plasmodiumvivax and Plasmodium cynomolgi as recognized by the natural anti-parasite

immune response. Molec Biochem Parasitol 89: 313–319.

32. Bonnet S, Petres S, Holm I, Fontaine T, Rosario S, et al. (2006) Soluble and

glyco-lipid modified baculovirus Plasmodium falciparum C-terminal merozoite

surface protein 1, two forms of a leading malaria vaccine candidate. Vaccine 24:

5997–6008.

33. Aribot G, Rogier C, Sarthou JL, Balde AT, Trape JF, et al. (1996) Pattern of

immunoglobulin isotype response to Plasmodium falciparum blood-stage

antigens in individuals living in a holoendemic area of Senegal (Dielmo, West

Africa). Am J Trop Med Hyg 54: 449–457.

34. Perraut R, Guillotte M, Drame I, Diouf B, Molez JF, et al. (2002) Evaluation of

anti-Plasmodium falciparum antibodies in Senegalese adults using different types

of crude extracts from various strains of parasite. Mic Infect 4: 31–35.

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 10 July 2014 | Volume 9 | Issue 7 | e101737

35. NGuer CM, Diallo TO, Diouf A, Tall A, Dieye A, et al. (1997) Plasmodiumfalciparum- and merozoite surface protein 1-specific antibody isotype balance inimmune Senegalese adults. Infect Immun 65: 4873–4876.

36. Perraut R, Mercereau-Puijalon O, Diouf B, Tall A, Guillotte M, et al. (2000)

Seasonal fluctuation of antibody levels to Plasmodium falciparum parasitized redblood cell-associated antigens in two Senegalese villages with different

transmission conditions. Am J Trop Med Hyg 62: 746–751.37. Perraut R, Morales-Betoulle S, Le Scanf C, Bourreau E, Guillotte M, et al.

(2000) Evaluation of immunogenicity and protective efficacy of carrier-free

Plasmodium falciparum R23 antigen in pre-exposed Saimiri sciureus monkeys.Vaccine 19: 59–67.

38. Rogier C, Trape JF (1993) Malaria attacks in children exposed to hightransmission: who is protected? Trans R Soc Trop Med Hyg 87: 245–246.

39. Genton B, Betuela I, Felger I, Al-Yaman F, Anders RF, et al. (2002) Arecombinant blood-stage malaria vaccine reduces Plasmodium falciparum

density and exerts selective pressure on parasite populations in a phase 1-2b

trial in Papua New Guinea. J Infect Dis 185: 820–827.

40. Fluck C, Schopflin S, Smith T, Genton B, Alpers MP, et al. (2007) Effect of the

malaria vaccine Combination B on merozoite surface antigen 2 diversity. Infect

Genet Evol 7: 44–51.

41. Thera MA, Doumbo OK, Coulibaly D, Laurens MB, Ouattara A, et al. (2011)

A field trial to assess a blood-stage malaria vaccine. N Engl J Med 365: 1004–

1013.

42. Garraud O, Mahanty S, Perraut R (2003) Malaria-specific antibody subclasses in

immune individuals: a key source of information for vaccine design. Trends

Immunol 24: 30–35.

43. Groux H, Gysin J (1990) Opsonization as an effector mechanism in human

protection against asexual blood stages of Plasmodium falciparum: functional

role of IgG subclasses. Res Immunol 141: 529–542.

PfMSP5 Antibodies Associated with Clinical Malaria Protection

PLOS ONE | www.plosone.org 11 July 2014 | Volume 9 | Issue 7 | e101737